Tag Archives: employment growth

The Impact on the Employment Numbers of the August 21 Announcement of the Bureau of Labor Statistics

Posted on by

A.  Introduction

The Bureau of Labor Statistics (BLS) issued an announcement on August 21 that said it had made a preliminary estimate that its figure for total employment as of March 2024 will be revised downwards by 818,000.  Some news media articles treated the announcement as if it were something to be alarmed by, and Trump issued a blast on the social media site he owns.  Trump asserted:  “MASSIVE SCANDAL!  The Harris-Biden Administration has been caught fraudulently manipulating Job Statistics to hide the true extent of the Economic Ruin they have inflicted upon America.  New Data from the Bureau of Labor Statistics shows that the Administration PADDED THE NUMBERS with an extra 818,000 Jobs that DO NOT EXIST, AND NEVER DID.  The real Numbers are much worse …” (sic, and capitalization as in the original).

None of this is true, but we know that accuracy has never been a strong point for Trump.  And such derogatory comments about the professionals at the Bureau of Labor Statistics just doing their jobs are also appalling.  There was nothing scandalous in their work.  A few basic points:

a)  Such a “preliminary benchmark revision” is issued every August, as part of an annual process by which the monthly employment estimates of the BLS are updated and anchored to (benchmarked to) more comprehensive estimates of employment.  This is done on a regular and routine basis every year.

b)  The date of the announcement is certainly not a secret, but rather is set well beforehand.  One will find it, for example, highlighted in a box on page 4 of the July jobs report that was released on August 2.  There was no attempt at a cover-up nor a leak.

c)  The 818,000 jobs figure is not some sort of monthly job number that people normally associate with the monthly jobs reports, but rather reflects an estimate of the change in the total number of people employed in March 2024.  The monthly employment estimates are then anchored to this benchmark, which will be updated again next year to an estimate for March 2025.  Employment still grew – and grew strongly – over the period from March 2023 (the previous benchmark) to March 2024 (which, when finalized, will become the new benchmark), but not by as much as was estimated before.  The previous estimate was of job growth of 2.9 million over this March to March period.  The new estimate (if the preliminary benchmark estimate holds – but bear in mind that it is preliminary and may well change) is of job growth of about 2.1 million.  That is still strong job growth.

d)  Many of the news articles highlighted that the 818,000 revision in estimated overall employment is high.  But one should keep in mind that it is equal only to about 0.5% of total employment.  That is, the revised figure (if the preliminary benchmark figure holds) will be 99.5% of what had been estimated earlier.  The 0.5% revision is also certainly not unprecedented.  Such revisions are part of a regular annual process, and figures the BLS provides going back to 1979 show that there have been revisions of 0.7% twice (in 1994 and 2009), 0.6% twice (1991 and 2006), and 0.5% four times (1979, 1986, 1995, and now in 2024).  That is, there have been such revisions to estimated overall employment by 0.5% or more a total of 8 times in 46 years, or 17% of the time.  A 0.5% change is large compared to what the figures normally are, but it is certainly not unprecedented, and in several years the revisions have been greater.

There is no scandal here.  There is no indication of manipulation.  And if there was some kind of politically motivated manipulation possible, doesn’t Trump realize that it would have made much more sense to manipulate the employment figures to be higher rather than lower?  Did he give even a few seconds of thought to his accusations?  The BLS is just doing the professional job it always has.

With all the publicity that has surrounded the BLS announcement, some may find of interest a description of how this annual updating process of the employment estimates works.  We will review that in the next section below.  The section following will then look at the figure itself – the 818,000 change in estimated overall employment – and what it may imply.  While still preliminary, the final estimate is likely to be close.  And the main message is that the basic story on employment growth during the Biden presidency has not changed.  Employment growth under Biden has been, and continues to be, exceptionally strong.

The chart at the top of this post updates a chart I provided in an article on this blog that was posted on August 21 – the day the BLS announcement came out.  I saw that announcement and the reports on it just after I posted that article.  One focus of that post was on the employment record under Biden and how it compared to the record under Trump.  The chart above replicates one in that August 21 post, but with the addition of what the path of estimated employment may now look like once the new benchmark is taken into account for the recent employment estimates.  That revised path is shown in orange.  It is a very rough estimate as the BLS has not yet worked out and released what the monthly employment figures will be with the new benchmark.  They are working on that now, and will release it – as they always do – in early February as part of the January monthly jobs report.

The path in orange is below the original one in red, but follows the same basic course.  It is still rising at a strong pace, and the basic message remains the same.  Job growth under Biden has been far stronger than what it was under Trump.

B.  The Annual Process of the BLS to Update Its Monthly Employment Estimates

The discussion in this section is based on material the BLS provides on its website on the process it follows in updating its monthly employment estimates to tie them (anchor them) to comprehensive employment estimates arrived at once a year from census-like figures.  The summary description provided here is based primarily on the BLS posts here and here.

The monthly jobs report of the BLS (more formally: “The Employment Situation” report) is eagerly awaited by many.  It provides estimates for what happened to the number of “jobs created” during the past month (more accurately, the change in the estimated number of nonfarm employees between the current month and the month before), as well as the unemployment rate along with numerous other measures of the labor market.

The report is produced on a very tight schedule.  The employment statistics come from a sample of establishments (both public and private, and called the Current Employment Statistics, or CES, survey), where the employing entities report to the BLS the number of employees on their payroll in the week of the month which includes the 12th day of the month.  The BLS jobs report is then issued at 8:30am on the Friday three weeks later, which is usually the first Friday of the following month.

(There are also figures in the monthly Employment Situation report on unemployment, the number in the labor force, and other figures that are obtained through the much smaller Current Population Survey (CPS) of households.  Most of what we will discuss here will be for the CES survey of business establishments, but similar modeling issues arise with the CPS survey, where there is also an annual process to update the model parameters.)

The survey of establishments is a rather comprehensive one, where the reporting entities account for about one-third of all nonfarm payroll jobs.  But it is still a sample survey, and the BLS needs to estimate from this survey the overall number of employees in the country (and hence what the change was from the previous month – the growth in the number employed).

For this, what is mainly needed is a large set of weights that the BLS can use to aggregate the reports it receives from firms of various types.  That is, to estimate the overall totals the BLS will need to know what weight to give to what is found in the survey reports for a particular type of firm (such as of a given size), operating in a particular sector, and perhaps categorized in other ways as well.

For example, small firms with up to 99 employees accounted for (in March 2023) 40.0% of all private employment in the country.  But while 70.4% of the number of private firms sampled by the BLS for the CES were in this category of up to 99 employees, those in the CES survey sample accounted for only 4.6% of total private employment.  Those firms are all small.  In contrast, large firms with employment of 1,000 or more were 6.2% of the number of private firms sampled by the BLS.  But those firms accounted for 68.4% of total private employment in the sample (and 28.8% of the total private employment in the country).

The BLS thus needs to know what weights to assign to each of these categories of firms to determine the overall totals.  The annual benchmarking exercise provides this.  A comprehensive census-type of exercise is needed, and for this the BLS uses primarily the March report of the Quarterly Census of Employment and Wages (QCEW) – which the BLS is also responsible for.  The QCEW is a comprehensive accounting of essentially all workers in the US based on the filings (and unemployment insurance tax payments) all firms are required to provide for the unemployment insurance program.

About 97% of the workers counted in the CES reports will be covered by regular unemployment insurance and hence included in the QCEW reports.  About 3% of workers are not, and the BLS uses various methods to arrive at a count for them.  Such “noncovered employment” (as the BLS labels it) includes, for example, certain workers at nonprofits and religious organizations, certain state and local government workers, railroad workers (where unemployment insurance is covered under the Railroad Retirement Board), paid interns and apprentices, and a range of others.

Keep in mind also that “employment” as reported in the monthly jobs report is for the nonfarm payroll, and thus excludes the self-employed as well as those working on farms (whether as self-employed owners or as employees).  But based on CPS data (the survey of households), those employed on farms (whether as employees or self-employed) only account for 1.4% of total employment.  That is so small that changes in on-farm employment do not have a significant impact on overall employment growth.  More potentially significant are the self-employed, who equal 6.1% of total employment according to the CPS data.  Unemployment insurance does not cover the self-employed, but those who are self-employed are also not employees and hence are not included in the CES definition of the nonfarm payroll.

The BLS then uses the detailed census counts from the March QCEW each year (supplemented by various sources of information for the remaining 3% of employees) to work out the weights to use to aggregate to the global estimates.  The March QCEW figures (as supplemented for the remaining 3%) then serve as an anchor on the employment totals.  It is updated on a routine basis annually on a calendar schedule that is set well ahead of time.  The monthly employment estimates are then worked out over the course of the year relative to the annual anchors of every March.

In addition to working out the weights to use to go from the monthly survey results to the overall totals, the BLS must also estimate the changes over time in the number of firms in each category.  That is, it needs to have an estimate for the number of new firms in each category that have begun operations each month (births), plus the number of firms that have ceased operations (deaths).  The QCEW census data will, by its nature, have nothing on the births and only outdated and now wrong information on the deaths.  The BLS updates its model of firm births and deaths each year as well, as part of its annual process of updating the benchmarks.

There has been speculation that the relatively large estimated reduction in estimated total employment of 818,000 in March 2024 may have been due in part to issues in the estimates of firm births and deaths.  There was an especially large jump in the number of new business establishments that opened in 2021 – a jump of 33% over what it was in 2020 or an increase of 37% over what it was in 2019 – to 1.4 million new firms in that year.  And the number of new firms was again at this record high of 1.4 million in 2022.  But small new firms typically struggle after a year or two, and many close even in the best of times.  It is possible that the BLS model for firm births and deaths did not capture well that this large jump in new business creation in 2021 and again in 2022 was followed by a relatively high number then closing in 2023 and 2024.

The BLS work begins once the March QCEW data become available, and each August it announces its preliminary benchmark revision for total employment in the prior March.  This is what the BLS announced on August 21, that Trump attacked.  The BLS will now work out the month-by-month implications of the new benchmark, adjusting the monthly employment figures that it had earlier estimated to reflect the new benchmark.  These revised monthly figures will be announced as part of the release of the January 2025 jobs report on Friday, February 7, 2025.  It does this in every January jobs report each year.

The benchmark figures on total employment are not seasonally adjusted numbers.  The anchors are the figures for each March, and hence the anchors in the upcoming revision will be for March 2023 (which is unchanged from what was determined before) and March 2024 (the new one).  The non-seasonally adjusted employment numbers will then be revised for the 21 months from April 2023 through to December 2024.  From April 2023 to the new March 2024 benchmark, the monthly employment figures will be adjusted in a simple linear fashion based on what the overall change in employment was between the March 2023 and March 2024 anchors.  If the final estimate turns out to be 818,000 (the same as the preliminary estimate), then that means the April 2023 non-seasonally adjusted employment estimate will be reduced by 68,167 (equal to one-twelfth of 818,000), the May 2023 estimate will be reduced by 136,333 (two-twelfths of 818,000), the June 2023 estimate by 204,500, and so on until the March 2024 employment estimate is reduced by 818,000.

The April 2024 to December 2024 figures for non-seasonally adjusted employment will then be re-estimated based on the models the BLS has updated based on the new March 2024 anchor estimates.  Keep in mind that by the time the January 2025 employment estimates are ready to be released (in early February 2025), the BLS will already have issued estimates for the April to December 2024 figures.  The revised estimates for all of the 2024 estimates are then provided in the Employment Situation report along with the employment figures for January.

The seasonally adjusted employment figures are then also updated.  Seasonally adjusted figures are calculated based on a statistical analysis of the regular annual patterns seen in the non-seasonally adjusted figures, using standard statistical programs.  The model parameters for this are re-estimated once the new non-seasonally adjusted employment figures are determined, and the BLS then goes back and revises the seasonally adjusted monthly employment estimates for a full five years.  Hence, once the January jobs report is released (on February 7 next year), one will find that the seasonally adjusted employment figures for the most recent five years (available online) will have also changed.

The January jobs report also has a section, in the interest of full transparency, showing what the new seasonally adjusted employment estimates are for each month of the past year, what the BLS had previously published, the difference, and the month-to-month employment changes (number of “new jobs”) as revised, as published before, and the difference.  All of this is routine.

The process is well-established and has been followed for at least 46 years (I have not looked farther back).  While the methods constantly evolve and are improved over time, there is no basis for Trump’s attack on the integrity of the BLS.

C.  How Much of an Impact?

The BLS was clear in its announcement that the new benchmark estimate for total employment in March 2024 is preliminary.  It is making this initial estimate available to the public in the interest of transparency, even though it has yet to work out the implications for the month-to-month employment figures.

But while preliminary with month-to-month specifics yet to be estimated, it is possible to get a sense of how significant a change this will likely entail to the pattern of employment growth under Biden.  And the answer is not much.  Furthermore, the change is in the direction one should have expected.  As discussed in my recent post on the economic record of Trump compared to that of Biden and Obama, employment growth during Biden’s term has been extremely fast.  This growth (whether based on the prior estimates or the preliminary revised estimates) has continued at a pace over the last year that is well in excess of separate estimates of growth in the labor force.  Over time, and at a constant unemployment rate, employment can only grow as fast as the labor force does.  In the past year the labor force participation rate rose slightly (from 62.6% of the adult population to 62.7%), which led to somewhat faster growth in the labor force than would be the case with a constant participation rate.  But the longer-term trend has been for the participation rate to drift downwards, as an aging population is leading to a higher share of adults in the usual retirement years.

The current estimate for the period of March 2023 to March 2024 – prior to any benchmark change – has been that total employment grew by 2.90 million.  This is based on the seasonally adjusted figures.  Growth over this period in the non-seasonally adjusted figures was a similar 2.96 million.  The preliminary benchmark change in total employment in March 2024 is 818,000, and formally this is the change in the non-seasonally adjusted figure for employment.  But it makes little difference whether one uses this to adjust the seasonally adjusted figures on employment or the non-seasonally adjusted figures.  With either, one ends up with a new figure for total employment in March 2024 of 2.1 million within round-off.

The month-by-month changes in the total employment estimates have yet to be worked out by the BLS, as noted before, but one can make a very rough estimate of what those might be.  The aim here is simply to give a sense of what the magnitudes are so that one can see – as in the chart at the top of this post – what the path of employment under Biden might then look like in comparison to the paths under Trump and Obama.

A number of assumptions are needed.  First, while the 818,000 adjustment in the benchmark employment total is formally a non-seasonally adjusted figure, I will assume the seasonally adjusted estimate will be similar.  The chart at the top of this post uses seasonally adjusted figures throughout, and the adjusted path for employment growth under Biden will be as well.

Second, for the period from April 2023 to March 2024 I adjusted the month-by-month employment estimates linearly, as the BLS does (although the BLS does this with the non-seasonally adjusted figures for the monthly employment estimates; I am assuming the changes in the seasonally adjusted figures will be similar).  That is, the April 2023 employment total was reduced by 68,167 (one-twelfth of 818,000), the May 2023 total by 136,333 (two-twelfths), and so on to March 2024.

Third, adjusting the figures going forward from March 2024 is more difficult as the BLS will use its updated models to make the revisions to the estimates from April.  Note that while the revised BLS estimates – when they are released as part of the January Employment Situation report – will cover the months through to December, all that we need now are estimates for the months of April, May, June, and July.

While very rough, for this I assumed the revisions for these four months will follow a pattern similar to what was found in the 2019 revision.  This was relatively recent but also pre-Covid (with all of the disruptions of patterns associated with that), and in that year the benchmark employment estimate was reduced by 0.3%.  While less than the 0.5% preliminary revision in the 2024 benchmark estimate, it was a still major revision downward (and during the Trump administration, although I do not recall ever seeing a reference by Trump to that reduction in the job totals).  I then used the month-by-month revisions in the seasonally adjusted employment estimates in 2019 for April through July, rescaled the percentage changes of each by the ratio of 0.5%/0.3% (in fact using the more precise figures of 0.517%/0.341%) and then applied those adjusted percentage changes to the current estimates of total employment in those four months.

The new path for total employment for the period of March 2023 to July 2024 is then shown as the orange line in the chart at the top of this post.  While below the current employment estimates (the line in red), the difference is not large.

The basic story remains the same.  Employment growth has been exceptionally strong under Biden, and has continued.  A downward revision in the benchmark total for March 2024 of 818,000 does not change this.

Trump’s Claims on the Economy and the Reality: A Comparison of Trump to Biden and Obama

Posted on by

“We had the greatest economy in the history of the world.  We had never done anything like it. … Nobody had seen anything like it.”

Donald Trump, Republican National Convention, Milwaukee, July 18, 2024

A.  Introduction

Donald Trump is fond of asserting that the US “had the greatest economy in the history of the world” while he was president.  He claimed this when he accepted the nomination at the Republican National Convention (as quoted above); he claimed it when he debated President Biden in June; and it is a standard line repeated at his campaign rallies.  He also asserts that this is all in sharp contrast to the economy he inherited from Obama and to where it is now under Biden.  In a June 22 speech, for example, Trump said “Under Biden, the economy is in ruins.”

These assertions of Trump are not new.  He was already repeatedly making this claim in 2018 – in the second year of his administration – asserting that the US was then enjoying “the greatest economy that we’ve had in our history” (or with similar wording).  And he repeated it.  The Washington Post Fact Checker recorded in their database that Trump made this claim in public fora at least 493 different times (from what they were able to find and verify) by the end of his term in January 2021.

Repetition does not make something true.  And numerous fact-checkers have shown that the assertion is certainly not true (see, for example, here, here, and here, and for the 2018 statements here).  But readers of this blog may nonetheless find a review of the actual data to be of interest, and in charts so that the extent to which Trump is simply making this up is clear.

The post will focus on Trump’s record compared to that of Obama’s second presidential term (immediately before Trump) and Biden’s presidential term (immediately after).  The post will also show that even if you just focus on the first three years of his presidential term – thus excluding the economic collapse in his fourth year during the Covid crisis – Trump’s record is nothing special.  The collapse in that fourth year was certainly severe, and with that included Trump’s record would have been one of the worst in US history.  But Covid would have been difficult to manage even by the most capable of administrations.  Trump’s was far from that, and that mismanagement had economic consequences, but Trump’s record is not exceptional even if you leave that fourth year out.

This post complements and basically updates a longer post on this blog from September 2020.  That post compared Trump’s economic record not only to that of Obama but also to that of American presidents going back to Nixon/Ford.  I will not repeat those comparisons here as they would not have changed.  I will focus this post on just a few of the key comparisons, adding in the record of Biden.

B.  The Record on Growth

The two charts at the top of this post show how Trump’s record compares to that of Obama and Biden in the two measures most commonly taken as indicators of economic performance – growth in national output (real GDP) and growth in total employment (jobs).  This section will focus on Trump’s not-so-special record on growth, while the section following will focus on employment.

Trump has repeatedly asserted that economic growth while he was president surpassed that of any in history.  This is not remotely true in comparison to growth under a number of post-World War II presidents.  (Quarterly GDP statistics only began in 1947 so older comparisons are more difficult, but there were certainly many other cases further back as well.)  Giving Trump the benefit of excluding the economic collapse in 2020 during the Covid crisis, real GDP grew at an annual rate of 2.8% over the first three years of Trump’s presidential term.  But real GDP grew at an annual rate of 5.3% during the eight years of the Kennedy/Johnson presidency; at a rate of 3.7% during the Clinton presidency; 3.4% during Reagan; and 3.4% as well during the Carter presidency.  The 2.8% during the first three years of Trump is not so historic.  Carter’s economic record is often disparaged (inappropriately), but Carter’s record on GDP growth is significantly better than that of Trump – even when one leaves out the collapse in the fourth year of Trump’s presidency.

Nor is the Trump record on growth anything special compared to that of Biden or Obama.  As seen in the chart at the top of this post, growth under Biden over the first three years of his presidency matched what Trump bragged about for that period (it was in fact very slightly higher for Biden).  GDP growth then remained strong in the fourth year of Biden’s presidency instead of collapsing.  Growth in the Obama presidential term immediately preceding Trump was also similar:  sometimes a bit above and sometimes a bit below, and with no collapse in the fourth year.  It was also similar in Obama’s first term once he had turned around the economy from the economic and financial collapse he inherited from the last year of the Bush presidency.

Trump’s repeated assertion that “we had the greatest economy in the history of the world” was a result – he claimed – of the tax cuts that Republicans rammed through Congress (with debate blocked) in December 2017.  While the law did cut individual income tax rates to an extent (heavily weighted to benefit higher income groups), the centerpiece was a cut in the tax rate on corporate profits from 35% to just 21%.  The argument made was that this dramatic slashing of taxes on corporate profits would lead the companies to invest more, and that this spur to investment would lead to faster growth in GDP benefiting everyone.

That did not happen.  As we have already seen, real GDP did not grow faster under Trump than it had before (nor since under Biden).  Nor, as one can see in the chart at the top of this post, was there any acceleration in the pace of GDP growth starting in 2018 when the new tax law went into effect in the second year of his presidential term (i.e. starting in Quarter 5 in the charts).

The promised acceleration in growth was supposed to be a consequence of a sustained spur to greater private investment from the far lower taxes on corporate profits.  There is no evidence of that either:

The measure here is of fixed investment (i.e. excluding inventories), by the private sector (not government), in real terms (not nominal), and nonresidential (not in housing but rather in factories, machinery and equipment, office structures, and similar investments in support of production by private firms).

This private investment grew as fast or often faster under Obama (when the tax rate on corporate profits was 35%) as under Trump (when the tax rate was cut to just 21%).  Growth under Biden has also been similar, even though the tax rate on corporate profits remains at 21%.  This similar growth is, in fact, somewhat of a surprise, as the Fed raised interest rates sharply starting in March 2022 with the aim of slowing private investment and hence the economy in order to bring down inflation.

With the far lower corporate profit tax rates going into effect in the first quarter of 2018 and the Fed raising interest rates starting in the first quarter of 2022 – both cases in the fifth quarter of the Trump and Biden presidential terms respectively – a natural question is what happened to private investment in the periods following those changes?  Rebasing real private non-residential fixed investment to 100 in the fourth quarter of the presidential terms, one has:

The paths followed by private investment under Biden (facing the higher interest rates of the Fed) and under Trump (following corporate profit taxes being slashed) were largely the same – with the path under Biden often a bit higher.  They diverged only in the 12th quarter of each administration (the fourth quarter of 2019 for Trump, and the fourth quarter of 2023 for Biden).  Under Trump, private investment fell in that quarter – well before Covid appeared – and then collapsed once Covid did appear.  Under Biden, in contrast, it kept rising up until the most recent period for which we have data.

It is also worth noting that private investment during the similar period in Obama’s second term rose by even more than under Trump (and for a period faster than under Biden, although later it rose by more under Biden).  This was despite a tax rate on corporate profits that was still at 35% when Obama was in office.  There is no evidence the tax rate mattered.  And although not shown in the chart here, private investment rose by far more in the similar period during Obama’s first term (although from a low base following the 2008 economic collapse).

With similar growth in such investment in all three presidential terms (leaving out the collapse in 2020), the conclusion one can draw is that taxes at such rates on corporate profits simply do not have a meaningful impact on investment decisions.  Decisions on how much to invest and on what depend on other factors, with a tax rate on profits of 21% or of 35% not being central.  Nor did the Fed’s higher interest rates matter all that much to investment during Biden’s term.  With a strong economy under Biden, firms recognized that there were investment opportunities to exploit, and they did.

The far lower tax rate of 21% on corporate profits did, however, lead to a windfall gain for those who owned these companies.  Far less was paid in such taxes.  That is, the tax cuts did have distributional consequences.  But they did not spur private investment nor overall growth.  They did not lead to “the greatest economy in the history of the world”.

C.  The Record on Employment

As seen in the chart at the top of this post, growth in total employment was higher under Obama than it was under Trump, and has been far higher under Biden – even if you restrict the comparison to the first three years of the respective presidential terms.  In the face of this clear evidence in favor of Biden’s record, Trump has now started to assert that the growth in jobs under Biden was due to a “bounce back” in jobs following the collapse in the last year of his administration, or that they all went to new immigrants.  But neither is true.

First, as one can see in the chart there has been strong growth in the number employed not only early in Biden’s administration but on a sustained basis throughout.  And second, nor was the growth only in the employment of immigrants.  The Bureau of Labor Statistics provides figures from its Current Population Survey (CPS) of households on the employment of those who were born in the US (the native-born) and those born abroad (the foreign-born).  Leaving out the collapse in 2020, employment growth over the first three years of Trump’s presidential term of the native-born averaged 1.3% per year.  During the first three years of the Biden presidential term, employment growth of the native-born averaged 1.8% per year.  The growth in employment of the native-born was not zero under Biden – as Trump claims – but rather was faster under Biden than under Trump.  While there is a good deal of noise in the CPS figures (which will be discussed below), these numbers do not provide support for Trump’s assertion.

There has also been concern expressed in the media with what was interpreted as a “disappointing” growth in employment in July.  The BLS “Employment Situation” report for July, released on August 2, indicated that employment rose by an estimated 114,000 in the month.  This is a good deal below the average in the 12 months leading up to July of 209,300 per month.  But an increase of 114,000 net new jobs in the month is substantial.  While there will often be large month-to-month fluctuations, one should not expect more on average going forward.

With the economy basically at full employment (the recent uptick in the unemployment rate – to a still low 4.3% – will be discussed below), the number employed cannot grow on a sustained basis faster than the labor force does.  And the labor force will grow at a monthly pace dictated by growth in the adult civilian population (i.e. age 16 and over) and what share of that adult population chooses to participate in the labor force.  The labor force participation rate in July was 62.7% and has been trending downward over the past several decades.  While a number of factors are behind this, the primary one has been the aging of the population structure with the Baby Boom generation moving into their normal retirement years.

The BLS report (using figures obtained from the Census Bureau) indicates that the adult civilian population rose by an average of 136,800 per month in the 12 months leading up to July.  At a labor force participation rate of 62.7%, the labor force would thus have increased by 85,800 per month.  Without an increase in the labor force participation rate, employment cannot grow faster than this on a sustained basis going forward.

In the past 12 months, however, the BLS report for July indicates that the labor force in fact grew at an average pace of 109,700 per month.  How was this possible?  The reason is that although the labor force participation rate is on a long-term downward trend due to the aging population, there can be and have been fluctuations around this trend.  And a small fluctuation can have a significant effect.  The labor force participation rate one year ago in July 2023 was 62.6%, and thus the rate in fact rose by 0.1% from July 2023 to July 2024.  If the labor force participation rate in July 2023 had in fact been 62.7%, then the labor force in July 2023 would have been 167,410,000 rather than the actual 167,113,000, and the increase over the 12 months leading to July 2024 would have averaged 84,900.  Within round-off, this is the same as the 85,800 figure calculated in the preceding paragraph for a constant 62.7% labor force participation rate,  (With more significant digits, the labor force participation rates were 62.589% and 62.696% respectively, and a constant 62.696% participation rate would have yielded the 85,800 figure for labor force growth.)

We should therefore not expect, going forward, that monthly employment will increase on a sustained basis by more than about 90,000 or so, or even less.  It could be higher if the labor force participation rate increases (and a small change can have a major effect), but the trend over the past couple of decades has been downward – as noted already – due to the aging of the population.  How then, was it possible for employment to have gone up by an average of 209,300 per month over the past year?  And this was also a period where the estimated unemployment rate rose from 3.5% in July 2023 to 4.3% in July 2024, which “absorbed” a share of the increase in the labor force as well.

The reason for these not fully consistent numbers is that employment estimates come from the Current Employment Statistics (CES) survey of establishments where people are employed, while the labor force and unemployment estimates come from the different Current Population Survey (CPS) – a survey of households.  The CES is a survey of nonfarm employers in both the private and public sectors, and covers 119,000 different establishments at 629,000 different worksites each month.  The “sample” (if it can be called that) covers an estimated one-third of all employees.

The CPS, in contrast, is a survey of about 60,000 households each month.  There will only generally be one or two members of the labor force in each household, so the share of the labor force covered will be far less than in the CES.  If each household had two members in the labor force, for example, the total of 120,000 would be only 0.07% of the labor force –  a sharp contrast to the one-third covered in the CES.  There is therefore much more statistical noise in the CPS data.  There are also definitional differences:  The CPS will include not only those employed on farms but also the self-employed and those employed in households.  Also, a person with two or more jobs will be counted as one person “employed” in the CPS.  The CES, in contrast, counts the employees of a firm, and the employers will not know if the individual may be working at a second job as well.  Thus a person working two jobs at two different firms will be counted as two “employees” in the CES.

These definitional differences are not major, however, and in part offset each other.  An earlier post on this blog looked at these differences in detail, and how, in an earlier period (2018/2019) there was a substantial deviation in the employment growth figures between the estimates in the CES and the CPS.  This was the case even with the figures adjusted (to the extent possible) to the same definition of “employment” in each.  There is a similar deviation between the employment estimates in the CES and in the CPS currently, with this accounting for a strong growth in employment as estimated by the CES (of 209,300 net new jobs each month over the past year) even though the labor force has grown -according to the CPS – by a more modest 109,700 per month over this period.

The labor market remains tight, however, even with the rise in the estimated unemployment rate to 4.3% in July:

The unemployment rate fell rapidly under Biden, following the chaos of 2020.  It was at a rate of 3.9% or less for over two years (27 months), despite the efforts by the Fed to slow the economy by raising interest rates.  The unemployment rate was also 3.9% or less for a period under Trump (for 20 months).  But as one sees in the chart, during the first three years of Trump’s term it basically followed the same downward path as it had under Obama.  It then shot up in March 2020 when the nation was caught unprepared for Covid.  As with the other key economic indicators (the growth in GDP, in employment, and in private investment), the paths followed by the economy during the first three years of Trump’s term were basically the same as – although usually not quite as good as – the paths set during Obama’s presidency.  They all then collapsed in Trump’s fourth year.

Any unemployment rate near 4%, and indeed near 5%, is traditionally seen as low.  Economists have defined the concept of the “Non-Accelerating Inflation Rate of Unemployment” (NAIRU) as the rate of unemployment that can be sustained without being so low that inflation will start to rise.  While one can question how robust this concept is (as will be discussed below), the NAIRU rate of unemployment has generally been estimated (for example by the staff at the Federal Reserve Board) to be between 5 and 6%.  An unemployment rate of 4.3% is well below this.  While the unemployment rate has gone up some in recent months, it is still extremely low.

D.  The Record on Real Living Standards

Ultimately, what matters is not the growth in overall output (GDP) or in employment, but rather in real living standards.  Many have asserted that because of recent inflation, living standards have gone down during Biden’s presidential term.  This is not true, as we will see below.  But first we will look at inflation.

Inflation rose significantly early in Biden’s presidential term.  The pace moderated in mid-2022, but until recently prices continued to rise:

Inflation was less during Trump’s term in office but was even lower under Obama.  Indeed, consumer price inflation has been low since around 1997, during Clinton’s presidency, until the jump in 2021.  Why did that happen?

The rise in 2021 can be attributed to both demand and supply factors.  On the demand side, both Trump and Biden supported and signed into law a series of genuinely huge fiscal packages to provide relief and support during the Covid crisis.  The packages were popular – especially the checks sent to most Americans (up to a relatively high income ceiling) that between the various packages totaled $3,200 per person.  But the overall cost for all the various programs supported was $5.7 trillion.  That is huge.  The funds were spent mostly over the two years of 2020 (under Trump) and 2021 (under Biden), and $5.7 trillion was the equivalent of 12.8% of GDP over those two years.  Or, as another comparison, the total paid in individual income taxes in the US in the single year of FY2023 was “only” $2.2 trillion.

While there was this very substantial income support provided through the series of Covid relief packages, households were limited in how much they could spend – out of both these income transfers and their regular incomes – in 2020 due to the Covid pandemic.  One went out only when necessary, and kept only to shopping that was necessary.  This carried over into early 2021.  But people could become more active as the Biden administration rolled out the massive vaccination campaign in the first half of 2021.  People then had a backlog of items to buy as well as the means to do so from what had been saved in 2020 and early 2021.  Demand rose sharply, and indeed Personal Consumption Expenditures in the GDP accounts rose by more in 2021 (by 8.4%) than in any year since 1946 (when it rose by 12.4%, and for similar reasons).

But at the same time, supply was constrained.  Supply chains had been sharply disrupted in 2020 worldwide due to Covid, and took some time to return to normal.  There was then the additional shock from the Russian invasion of Ukraine in February 2022, leading oil and many other commodity prices to spike.

Supply chains did, however, return more or less to normal early in the summer of 2022.  And as they did, one saw a sudden and sharp reduction in pressures on prices, in particular on the prices of goods that can be traded:

This chart shows the annualized inflation rates for 6-month rolling periods (ending on the dates shown) for the overall CPI, for the shelter component of the CPI, and for the CPI excluding shelter.  The overall inflation rate rose from an annualized rate of 3.2% in the six months ending in January 2021 (the end of Trump’s term) to a peak of 10.4% in the six months ending in June 2022.  It then fell remarkably fast, to an annualized rate of just 2.6% in the six months ending in December 2022.

This sudden drop in the inflation rate is seen even more clearly in the CPI index of prices for everything but shelter:  The annualized rate fell from 12.4% in the first half of 2022 (the six months ending in June) to a negative 0.2% rate in the second half of 2022 (the six months ending in December).  Why?  There was not a sudden collapse in consumer or other demand.  Rather, supply chains finally normalized in the summer of 2022, and this shifted pricing behavior.  When markets are supply constrained (as they were with the supply chain problems), firms can and will raise prices as competitors cannot step in and supply what the purchaser wants – they are all supply constrained.  But as the supply chains normalized, pricing returned to its normal condition where higher demand can be met by higher production – whether by the firm itself or, if it is unwilling, by its competitors.  It is similar to a phase change in conditions.

Shelter is different.  It covers all living accommodations (whether owned or rented), and as has been discussed in earlier posts on this blog (see here and here), the cost of shelter is special in the way it is estimated for the CPI.  It is also important, with a weight of 36% in the overall CPI index (and 45% in the core CPI index, where the core index excludes food and energy).  The data for the shelter component of the CPI comes from changes observed in the rents paid by those who rent their accommodation, and rental contracts are normally set for a year.  Hence, rental rates (and therefore the prices of the shelter component of the CPI) respond only with a lag.  One can see that in the chart above, with the peak in the inflation rate for shelter well after the peak in the inflation rate for the rest of the CPI.

Since mid-2022, the rate of inflation as measured by the overall CPI has generally been in the range of 3 to 4% annualized.  Increases in the cost of shelter have kept it relatively high and above the Fed’s target of about 2% per annum.  But as seen in the chart, it has recently come down – falling to an annualized rate of 2.5% in the six months ending in July.  For everything but shelter, the rate in the six months ending in July was only 1.4%.

One question that some might raise is whether the very tight labor markets – with an unemployment rate that was 4% or less until two months ago – might have led to the inflation observed.  The answer is no.  As noted above, inflation in all but shelter fell suddenly in mid-2022, falling from a rate of 12.4% in the first half of the year to a negative 0.2% in the second half, even though the unemployment rate was extremely low at 4% or less throughout (and only 3.5 or 3.6% in all of the second half of 2022).  Unemployment has remained low since while inflation has come down.  If the cause was tight labor markets, then the rate of inflation would have gone up rather than down.

Similarly, inflation as measured by the CPI was not high in 2018 nor in 2019 when labor markets were almost as tight during Trump’s presidency – with overall inflation then between 2 and 3% on an annual basis.  Nor did inflation go up during the similarly tight labor market of 1999 and 2000 during the Clinton presidency:  CPI inflation was generally in the 1 1/2 to 3 1/2 % range during that period.  All this calls into question the NAIRU concept, with its estimate that an unemployment rate below somewhere in the 5 to 6% range will lead to pressures that will raise the rate of inflation.

Managing inflation coming out of the chaos of 2020 was certainly difficult.  Inflation spiked in most countries of the world following the Covid crisis, reaching a peak in 2022.  But the rate of inflation has since come down as supply conditions normalized.  That does not mean that the absolute level of prices came down, only that they were no longer increasing at some high rate.  Wages and other sources of income will then adjust to the new price levels, and what matters in the end is whether real levels of consumption improve or not.  And they have:

The chart shows the paths followed for per capita real levels of personal consumption expenditures, as measured in the GDP accounts, during the presidential terms of Trump, Biden, and the second term of Obama.  The path followed under Trump was basically the same as that followed under Obama – until the collapse in the last year of Trump’s term.  The path followed under Biden has been substantially higher than either.  It was boosted in his first year as the successful vaccination campaign allowed people to return to their normal lives.  They could then purchase items with not only their then current incomes, but also with the savings they had built up in 2020.  But even if one excludes that first year, the growth under Biden has been similar to that under Obama and under Trump up to the collapse in Trump’s fourth year.

Once again, there is no basis for Trump’s claim of the “greatest economy”.

E.  Summary and Conclusion

The economy during Trump’s presidency was certainly not “the greatest in the history of the world”.  Nor was it even if you leave out the disastrous fourth year of his presidency.  Covid would have been difficult to manage even by the most capable of administrations, and Trump’s was far from that.  Instead of preparing for the shock this highly contagious disease would bring, Trump’s response was to insist – repeatedly – “it’s going to go away”.

Trump’s economic record was certainly nothing special.  Real GDP grew as fast or faster under Obama and Biden as it had under Trump.  Trump insisted that growth would be – and was – spurred by the tax cuts that he signed into law in late 2017 that slashed the tax on corporate profits.  But there is no indication of this in the data.  Nor is there even any indication that private investment rose as a result of the lower taxes.

Employment has grown far faster under Biden than it had under Trump, and also grew faster in Obama’s second term – even leaving out Trump’s disastrous fourth year.  Unemployment fell during the first three years of Trump’s term in office (before sky-rocketing in his fourth year), but here it just followed a very similar path to that under Obama.  For this, as with GDP and employment growth, perhaps the biggest accomplishment of Trump’s first three years in office was that he did not mess up the path that had been set under Obama.  And unemployment has been even lower under Biden.

Inflation was certainly higher in 2021 as the US came out of the Covid crisis.  Supply chains were still snarled, but there was pent-up demand from consumers who had had to avoid shopping in 2020 due to Covid and who also benefited from a truly huge set of Covid relief packages passed under both Trump and Biden.  Supply chains then normalized in mid-2022, sharply reducing pricing pressures for goods other than shelter.  Due in part to lags in how rental rates for housing are set (as they are normally fixed for a year) and then estimated by the BLS, the cost of the shelter component of the CPI came down more slowly than the cost of the rest of the CPI.  This kept inflation as measured higher than what the Fed aims for, although recently (in the last half year) it has come down again.  Most anticipate that the Fed will soon start to cut interest rates from their current high levels.  The inflationary episode resulting from the Covid crisis appears to be coming to an end.

There is thus no justification for the claim by Trump that “we had the greatest economy in the history of the world”.  Yet he has repeatedly asserted it, both now and when he was president.  Why?  Stephanie Grisham, who served in the Trump administration as press secretary and in other senior positions, and who had been – by her own description – personally close to Trump, explained it well in a speech she made on August 20 to the Democratic National Convention.  She noted that Trump used to tell her:  “It doesn’t matter what you say, Stephanie.  Say it enough, and people will believe you.”

Many do appear to believe that the economy was exceptionally strong when Trump was president:  that it was “the greatest in history”.  But that is certainly not true.  Facts matter; reality matters; and a president needs to know that they matter.

Raising the Minimum Wage Has Not Led to Higher Unemployment: Evidence from California

Posted on by

A.  Introduction

California has aggressively increased its minimum wage since 2014, starting on July 1 of that year and then with increases on January 1 of each year from 2016 through to 2024.  Critics have argued that this would increase unemployment, saying that firms would no longer be willing to employ minimum-wage workers at the new, higher, minimum wage rates.  They argued that the productivity of these workers was simply too low.  If they were right, then one would have seen increases in the unemployment rate in the months following each of the steps up in the minimum wage.  But there is absolutely no evidence that this happened.

The chart at the top of this post shows this lack of a response graphically.  It may be a bit difficult to see as showing a lack of a response is more difficult than showing the presence of a response.  The chart will be discussed in more detail below, but briefly, it shows the averages in each of the subsequent 12 months following the increases in the California minimum wage (including or excluding 2020 to 2022, as the Covid disruptions dominated in those years), of the change in the unemployment rate in California versus the change in the unemployment rate in the US as a whole.  The changes are defined relative to what the unemployment rates were in the month before the increase in the minimum wage – i.e. the comparison is normally to the rate in December when the new minimum wage became effective on January 1.  The unemployment rate of course goes up and down depending on macro conditions (and was normally going down for most of this period), so to control for this the changes in the unemployment rate in California are defined relative to the changes in the US as a whole.

What was the result?  The chart shows that basically nothing happened.  If anything, what was most common was that the unemployment rate fell slightly in California relative to the rate in the US in the months following increases in the California minimum wage.  These changes were small, however, and are not really significant.  But what is clear and significant is that aggressive increases in the minimum wage in California have not led to increases in unemployment in the state.  The assertion that they would is simply wrong.

As noted above, this chart will be discussed in more detail below.  But the post will first look at the changes in the minimum wage in California since 2014, and how the minimum wage in California compared to the federal minimum wage for the US as a whole as well as to several measures of wages in the US and to the federal poverty line.  Following a look at the (non)-impact on unemployment, we will for completeness also examine what happened to labor force participation rates.  Some might argue that minimum-wage workers who would have lost their jobs might then have left the labor force (in which case they would not have been counted as unemployed).  But we will see that labor force participation rates in California also did not change following increases in the minimum wage.  Finally, the post will discuss possible reasons for why increases in the minimum wage in California did not lead to a rise in unemployment there.  Standard economics under the standard assumptions would have predicted that it would have.  But those standard assumptions do not reflect well what is happening in the real world in labor markets.

B.  The Minimum Wage Rate in California

The federal government sets a minimum wage that applies to the US as a whole.  But due to gridlock in Congress (and opposition by Republicans), the last time the federal minimum wage was raised was in July 2009, when it was set at $7.25 per hour.  As was discussed in a post on this blog from 2013, when adjusted for inflation this minimum wage was below what we had in the Truman administration in 1950, despite labor productivity now being more than three times higher than then.  And from July 2009 to now, inflation has effectively reduced the value of the $7.25 wage of July 2009 to just $4.97 (based on the CPI).  The federal minimum wage has simply become irrelevant.

Due to this lack of action at the federal level. many states have legislated their own minimum wage rules for their respective jurisdictions.  California is one, and has been particularly aggressive.  Over the past decade, the minimum wage in California has been increased to $16 per hour generally and most recently to $20 per hour for fast-food restaurant workers:

California Minimum Wage Recent History

Effective date 25 employees or less 26 employees or more
Jan 1, 2008 $8.00 $8.00
July 1, 2014 $9.00 $9.00
Jan 1, 2016 $10.00 $10.00
Jan 1, 2017 $10.00 $10.50
Jan 1, 2018 $10.50 $11.00
Jan 1, 2019 $11.00 $12.00
Jan 1, 2020 $12.00 $13.00
Jan 1, 2021 $13.00 $14.00
Jan 1, 2022 $14.00 $15.00
Jan 1, 2023 $15.50 $15.50
Jan 1, 2024 $16.00 $16.00
Fast food restaurant employees:
Apr 1, 2024 $20.00 $20.00

Sources:  California Department of Industrial Relations.  See here and here.

The focus of this post is on the series of increases that began on July 1, 2014, with the prior minimum wage set as of January 1, 2008, shown for reference.  That 2008 rate was $8.00 per hour and was raised effective on July 1, 2014, to $9.00 per hour.  California then began to increase the minimum wage annually starting January 1, 2016, with this continuing up to and including on January 1 of this year (2024).  Furthermore, effective January 1, 2017, California began to set separate minimum wage rates for workers employed in businesses with 25 employees or less or with 26 employees or more.   These could differ, although recently they have not.

Finally and most recently, California set a new minimum wage effective on April 1, 2024, of $20 per hour for employees of fast food restaurants (in restaurant chains with 60 or more locations nationwide).  I include this here for completeness, but it is still too early to say whether this has had an impact on unemployment.  So far it has not, but as I write this state-level unemployment data is available only for the months of April and May.  But those figures do not provide any support for the critics:  The unemployment rate in California in fact fell in those two months compared to that in the US.  This will be discussed below.

The general California minimum wage has now doubled – to $16 per hour – from the $8 per hour it was prior to July 1, 2014.  But for a sense of what this means, it is useful to put this in terms of various comparators:

California Minimum Wage:  Selected Comparisons

California minimum wage in firms with 26 employees or more

California Minimum Wage per hour Ratio to US median wage of hourly workers Ratio to US average hourly earnings of all private sector workers Ratio to Poverty Line for family of four Ratio to upper limit of earnings of first decile of US wage & salary workers
2008 $8.00 65% 38% 76% 93%
2014 $9.00 68% 37% 76% 94%
2016 $10.00 71% 39% 83% 102%
2017 $10.50 72% 40% 86% 103%
2018 $11.00 73% 41% 89% 104%
2019 $12.00 78% 43% 94% 109%
2020 $13.00 79% 46% 100% 111%
2021 $14.00 82% 47% 107% 115%
2022 $15.00 83% 47% 109% 113%
2023 $15.50 81% 47% 104% 108%
2024 $16.00 46% 104% 108%
Fast Food:
April 2024 $20.00 58% 129%

The comparisons here are based on the California minimum wage for employees in businesses with 26 or more employees.

The wage measures come from various reports produced by the Bureau of Labor Statistics (BLS).  The first column (following the column with the California minimum wage) shows the ratio of that minimum wage to the BLS estimate of the US median hourly earnings of wage and salary workers paid an hourly wage.  The ultimate source for this is the Current Population Survey (CPS) of the BLS, and this particular series is only provided annually (with 2023 the most recent year).  The California minimum wage rose from 65% of this median wage of hourly workers in 2008 to 83% in 2022 and 81% in 2023).  By this measure of wages – of wage and salary workers paid an hourly wage – the California minimum wage rose significantly in comparison to what a median hourly worker was being paid nationally.

A broader measure of wages is provided in the next column.  The ratios here are for a worker being paid the California minimum wage to the average hourly earnings of all private sector workers – not just workers paid at an hourly rate.  This is also provided by the BLS, but comes from its Current Employment Statistics monthly survey – a survey of business establishments that asks firms how many they employ and what they were paying those workers.  These average wages are higher as they cover all workers and not only those paid at an hourly rate, plus the average will be higher than the median in cases such as this (as the distribution of wages paid is skewed to the right).  By this measure, the California minimum wage rose from 38% of what US private sector workers were being paid on average in 2008 (and 37% in 2014) to 46-47% since 2020.

In terms of the federal poverty line, even full-time workers (40 hours per week for 52 weeks each year) paid the minimum wage in California in 2008 or even 2014 would have been able to earn only 76% of the poverty line income for a family of four.  But with the increases in the minimum wage in the past decade, they would have finally been able to reach that poverty line in 2020, and then 109% of it in 2022.  In 2023 and again in 2024, it would have been 104%.

The final column shows earnings at the California minimum wage compared to the earnings that would place a worker in the first decile (the bottom 10%) of the distribution of earnings of full-time wage and salary workers.  These are also estimates from the BLS, are expressed in terms of usual weekly earnings, and are issued quarterly based on results from the CPS surveys.

With the increases in the California minimum wage over the past decade, full-time workers earning the minimum wage in California had incomes that exceeded the upper limit of the earnings of wage and salary workers in the US as a whole who were in the first decile of the earnings distribution – ranging from 102% of what the bottom 10% earned in 2016 to 115% in 2021 and 108% currently.  Assuming the distribution of earnings in California would be similar to that in the US in the absence of the special California minimum wage laws, this can provide a rough estimate of how many workers were being affected by the California minimum wage laws.

If earnings at the California minimum wage would have matched the earnings at the upper limit of the first decile (i.e. a 100% ratio), the implication would be that the share of workers for which the California minimum wage was applicable would be 10%.  With the ratio above 100% (by varying ratios up to 115%) the share affected would have been somewhat more than 10% – perhaps 11 or 12% of workers as a rough guess.  But the BLS data is not for the entire labor force.  Rather, it is only for wage and salary workers employed full-time.  One has, in addition, part-time workers and those who are self-employed.  The distribution of hourly earnings among those workers is not available, but if it is similar to the hourly earnings of full-time workers, the share affected would be the same 10% (or more).

The purpose here is just to provide a general feel for how many minimum wage workers were being affected by the changes enacted in the California minimum wage over the past decade.  Various factors cannot be accounted for, but they are at least in part offsetting.   For the purposes here, a reasonable estimate would be that at least 10% of the labor force had wages so low that the increases in the minimum wage in California over the last decade had an impact on what they would then be paid.  That is a not insignificant share.

C.  The Impact of Increases in the Minimum Wage on Unemployment

What impact did those increases in the California minimum wage then have on the employment of workers who were being paid the minimum wage? Critics of the minimum wage argue that workers are paid a wage based on their productivity, and if they are being paid at or close to the minimum wage this is only because their productivity is low.  In this view, if the minimum wage that has to be paid is then raised, those workers will be let go and will become unemployed.  Did we see this?

No, we did not.  The evidence from the ten different increases in the minimum wage in California over the past decade (from July 2014 to January 2024) does not show any impact at all on unemployment.  The chart at the top of this post summarizes the results.

The chart is based on calculations using data on the unemployment rate in California and on the unemployment rate in the US as a whole, where I calculated the unemployment rates from underlying data on the number unemployed and the number in the labor force (as published unemployment rates themselves are shown only to the nearest 0.1% point – anything less is not considered significant).

For numerous structural reasons, the unemployment rate in any particular state (including California) will differ from the rate in the nation as a whole.  These structural reasons include the age structure of the population (middle-aged workers are less likely to be unemployed than young workers), the education structure (college-educated workers are less likely to be unemployed than workers with only a high school education), the industrial structure, the racial and ethnic mix of the population, and much more.

But while these structural factors affect the level of the unemployment rate in California relative to the national average, such structural factors change only slowly over time and hence do not have a significant impact on the month-to-month changes in that rate.  The rate of unemployment itself can, however, change significantly from month to month at the national (as well as state) levels due to macroeconomic factors.  In a recession the rate of unemployment goes up, and in a recovery or during periods of rapid growth, the rate of unemployment goes down.  It is just that in the absence of some state-specific event (such as – possibly – a change in its mandated minimum wage), the month-to-month changes in the unemployment rate at the state level will generally be similar to the changes seen at the national level.  They move together, as affected by macroeconomic factors.  The question being examined is thus whether the increases in the minimum wage in California over the past decade led to an increase in the unemployment rate in California in the months following those changes in the minimum wage, as compared to what was observed for the unemployment rate nationally.

This is a simple form of what is called the “difference-in-difference” method.  What is significant is not whether unemployment in California went up or down during the period, but whether it went up or down by more than what was seen at the national level in the same period.  For example, define the changes as relative to the month prior to a change in the minimum wage law (i.e. normally relative to what the rate was in December, as all but one of the changes were effective on January 1 of each year).  The employment and unemployment statistics (gathered by the BLS as part of the CPS household surveys) take place in the middle week of each month, so the mid-January unemployment rate will be treated as month one following the change in the minimum wage.  The mid-February unemployment figures will then be month two, and so on until mid-December of that year will be month twelve.  The minimum wage was then increased again in the next January 1, and the annual cycle was repeated for a second set of observed impacts (or non-impacts).  The changes in the unemployment rate are thus defined as the difference between changes in the California rate for the given number of months following the change in its minimum wage (i.e. in month one, or in month two, and so on to month twelve), relative to what the changes were in the same period for the US as a whole.

As a concrete example using made-up numbers, suppose that in some December the unemployment rate in California was 6.0% while the unemployment rate in the US as a whole was 5.0%.  Suppose then that in, say, month three (March) the observed unemployment rate in the US was 4.5% – a fall of 0.5% point over the period.  If the unemployment rate in California fell to 5.5% in the same period (to March), then the change in California was the same as the change in the US as a whole, and the increase in the minimum wage on January 1 did not appear to have any differential effect.  If, however, the unemployment rate in California fell only by, say, 0.3% points to 5.7%, while the US rate fell by 0.5% in the same period, one would say that it appears the increase in the minimum wage in California led to an increase in its unemployment rate by 0.2% points.  And if the rate in California fell by 0.7% points to 5.3% while the US rate fell by 0.5%, then there was a 0.2% point reduction in the unemployment rate in California following the change in its minimum wage rate.

There will of course be statistical noise, as all the figures are based on household surveys.  And importantly, in any given year there will also be special factors that could enter in that particular year that could affect the results.  More is always happening than just a change in the minimum wage law.  But to address this we have that California changed its minimum wage law on ten separate occasions over this ten-year period.  We therefore have ten separate instances, and we can work out the average over those ten separate episodes.  While special factors may have arisen in any given year, the only common factor in all ten was that California raised its minimum wage ten separate times.

(The exception in the averages is for the January 1, 2024, increase in the minimum wage,  As I write this, we only have data for the five months through May.  Thus the averages over up to the full ten instances can only be calculated for the first five months, while the averages for months six through twelve can only be for the nine cases to 2023.  Also, note that for the July 1, 2014, increase in the minimum wage, the changes were defined relative to the California and US unemployment rates in June, with the subsequent twelve months then covering July 2014 to June 2015.)

Those average impacts were then remarkably small:

Average Changes in the California Unemployment Rate less Changes in the US Unemployment Rate, in the Months Following an Increase in the California Minimum Wage (in percentage points)

Months from Minimum Wage Change July 2014 –   May 2024 July 2014-2019,                   and 2023 – May 2024
0 0.00% 0.00%
1 -0.02% 0.01%
2 -0.06% -0.05%
3 -0.05% -0.04%
4 -0.07% -0.05%
5 0.00% -0.10%
6 -0.00% -0.09%
7 0.03% -0.08%
8 0.04% -0.10%
9 -0.04% -0.05%
10 -0.02% -0.05%
11 -0.02% -0.05%
12 0.02% -0.03%
Overall average -0.02% -0.06%

The chart at the top of this post shows this table graphically.  The two columns are for averages over the full period and with the years 2020 to 2022 excluded.  The Covid disruptions dominated in those years, but the results are basically the same whether those years are included or excluded.

The changes were all essentially zero.  It is not possible to see any increase in the California unemployment rate at all resulting from the increases in the minimum wage in the state over the past decade.  If anything, the increases in the minimum wage were associated in most cases with a small reduction in the unemployment rates.  But these are all small, and are probably simply statistical noise and not significant.

To put this in perspective, recall the discussion above that arrived at the rough estimate that the share of the labor force being paid at or close to the minimum wage might be around 10%, and possibly more.  If – as the critics argue – such workers can be paid only those low wages because their productivity is so low, then they would all lose their jobs if their employers were required to pay them a higher wage.  If true, the unemployment rate would then shoot up by 10% points.  One obviously does not see that.

If we had over-estimated the share employed at the minimum wage by a factor of two, so that it was in fact 5% rather than 10% of the labor force, then the unemployment rate would have shot up by 5% points.  One does not see that either.  One does not even see an increase of 1% point, nor, for that matter, even 0.1%.  The overall average change is in fact generally a small decrease in the rate of unemployment in California relative to the US rate in the months following an increase in the minimum wage, although I suspect this is just statistical noise.

Most recently, California raised the minimum wage for workers at fast food restaurants (at chains with 60 or more locations nationally) to $20 per hour effective April 1, 2024.  We so far only have data for April and May as I write this, but that data provides no support for the belief that this has led to an increase in the unemployment rate.  Fast-food workers are of course only a small share of the labor force:  about 2.2% in California in 2023 based on BLS data for fast-food and counter workers (where fast-food workers make up about 80% of this total in national data).  But in the two months since the April 1 increase to $20 per hour for fast food workers, the California unemployment rate relative to that in the US in fact fell by 0.06% points in April compared to March, and by 0.25% in May compared to March.  It did not go up but rather went down.

Finally, it is possible that critics of the minimum wage may argue that low-wage workers laid off following an increase in the minimum wage will then leave the labor force entirely.  If they did this, they would then not show up in the unemployment statistics and one would not see an increase in the observed unemployment rates.  To be counted as unemployed in the BLS surveys, the unemployed person must have taken some positive action in the prior four weeks to try to find a job (e.g. send out applications, visit an employment center, and similarly) and yet was not employed at the time of the survey.  If they did not take such an action to try to find a job, they would not be counted as “unemployed”.  Rather, they would be counted as not participating in the labor force.

Therefore, for completeness, I calculated what happened to the Labor Force Participation Rate in California compared to the US rate in the months following the increases in the California minimum wage.  The data comes from the BLS (but is most conveniently accessed via FRED, for the US and the California rates respectively):

Average Changes in the California Labor Force Participation Rate less Changes in the US Labor Force Participation Rate, in the Months Following an Increase in the California Minimum Wage (in percentage points)

Months from Minimum Wage Change July 2014 –  May 2024 July 2014-19,                       and 2023 – May 2024
0 0.00% 0.00%
1 -0.01% -0.06%
2 -0.02% -0.10%
3 -0.07% -0.11%
4 0.04% -0.10%
5 0.01% 0.01%
6 0.11% 0.00%
7 0.06% -0.08%
8 -0.02% -0.03%
9 -0.11% -0.10%
10 -0.11% -0.05%
11 -0.04% -0.05%
12 0.02% 0.02%
Overall average -0.01% -0.05%

As with the unemployment rates, there was no significant impact.  Had the 10% of the workers being paid at or close to the minimum wage dropped out of the labor force following the increases in the minimum wage, the figures would have shown a 10% point reduction in the California labor force participation rate.  One does not see anything remotely close to that.  One does not see an impact of even 1.0% point.  There was simply no significant impact on labor force participation rates.

Thus, the data indicates the minimum-wage workers remained in the labor force and did not become unemployed.

D.  The Economics of How Wages are Determined:  In Theory and in the Real World

Economic analysis, when done well, will be clear on what conditions are necessary for certain propositions to hold.  Under those conditions, one might be able to arrive at interesting conclusions.  But a good analyst will examine whether there is reason to believe that those conditions reflect what we should expect in the real world.  Often they do not.  That is, what is of interest is not simply some proposition in isolation, but rather also under what conditions one can expect that proposition to hold.

The economics of how wages are determined is a good example of this approach.  One can show that, under certain conditions, the wages paid to a worker would reflect the value of the marginal product of that worker – that is, the value of the increase in output that was made possible by hiring that worker.  But one should then look at the conditions that are necessary for this to follow.  And in the case of wage determination, they are not at all realistic, particularly for low-wage workers.  The implication is that one should not expect the wages of these workers to reflect necessarily the value of the marginal product of such a worker.

A problem, however, is that some commentators do not follow through and examine the conditions necessary for the theoretical conclusion to hold.  That is, they stop at the proposition that workers will be paid the value of their marginal product, and fail to look at whether the conditions under which that proposition would hold are realistic.  They thus conclude, for example, that increases in the minimum wage will lead to the layoff of all the workers who were being paid the prior minimum wage.  In their world, those workers are being paid a low wage because their productivity is low, and if firms are then required to pay a higher wage then those workers – these analysts conclude – will be laid off and indeed not be employable anywhere.  They assert that their productivity is too low.

Yet as we saw above, we see nothing at all close to this in the data.  California raised its minimum wage repeatedly in the last decade, and in a significant and meaningful way.  We saw that it led to a significant increase in the wages of such workers compared to the overall wage structure in the US.  Yet the unemployment rate in California did not increase at all in the months following those increases.

What, then, are the conditions that are necessary for this theoretical model of wage determination to hold?  And how realistic are they?  This section will provide a brief discussion of that theoretical model, and will then examine some of the conditions necessary for it to hold.  It will not be a comprehensive discussion of all the issues that could arise.  There are others as well.  Rather, the purpose is to show for one set of reasons (there could be others also), the simple notion that wages will be equal to the value of the marginal product of the worker does not reflect the reality of how wages are determined.

a.  The Standard Neoclassical Model of Wage Determination

In the standard model of neoclassical economics, it can be shown that the wages of a worker will equal the value of the marginal product of the worker.  This can be shown to hold under the assumption of “perfectly competitive markets” for both labor (hired as an input) and for firms (hiring the labor).  But for such perfectly competitive markets to exist, one needs:

1.  On the side of the firms, there are many firms within a small geographic zone (small enough that commuting costs to the firms will not differ significantly) that are all competing with each other to hire labor with any given skill set.  That is, the markets are “dense”, with many firms competing for that labor.

2.  On the side of labor, there are many workers with each given skill set who are competing with each other and are seeking to be employed within that geographic zone.

3.  There are no lumpy fixed costs incurred by the firms in hiring or firing a worker, nor are there any lumpy fixed costs for a worker in finding and being hired into a new job.  Economists refer to this as no transaction costs.  That is, that there are no costs incurred (neither on the part of the firm nor the worker) when a worker is fired and replaced with another.

4.  There is full information freely available to all parties on what skills are required for a job, what skills each worker has, and how any worker will perform in any job.  Both the firms and the workers know all this, with no cost to obtain such information.

5.  Production is a smooth, upwardly rising (up to some limit), and always concave function of the hours any individual laborer provides for a job.  Concave means that while the curve is rising, it is rising by less and less as the hours provided by the laborer increases.  That is, there are no “bumps” in the curve.  The slope of that curve at any given number of hours of labor is the marginal product of the laborer at that number of hours.  That is, the slope indicates how much additional output there will be with one additional unit of labor being provided.

If all of the above holds, then one should expect that firms will pay in wages, and workers will receive, the value of the marginal product of what the workers produce.  If workers were paid less than this, they would know the value of what they produce is in fact more and they would immediately move to a nearby competing firm that is willing to pay them up to the value of their marginal product.  And if firms paid more than this, then competing firms could take away business from the firms paying the higher wages.

In this system, workers will thus be paid the value of their marginal product – no more and no less.  And if this were true in the real world, then a mandate from the government to pay a higher minimum wage would mean that all those workers whose productivity was below the new minimum wage rate would be let go.  They would become unemployed and indeed unemployable, as this set of assumptions implies that the productivity of such workers is simply too low for any firm to be willing to pay them the new minimum wage.

b.  But the real world differs

Laying out the assumptions necessary for the neoclassical theory of wage determination allows us then to see whether those assumptions correspond to what we know about the world.  They do not:

1. Markets are rarely dense.  There are usually only a few firms – and often even no other firms hiring workers with similar skills – within a geographic zone so small that a worker is indifferent as to whom they would go to work for.  There may be few or even no firms nearby that a worker could threaten to move to if they are being underpaid.  And the few firms that are there may well follow what they consider to be informal “norms” on what such workers should be paid, rather than compete with each other and bid up the local wages.

2.  There are transaction costs for both a firm considering to fire a worker and then to hire a new worker as a replacement, and for a worker when considering a move to a new employer.  There are major costs incurred by both.  Switching between employers is far from cost-free, so it is rarely done.

3.  There can also be more overt constraints imposed on labor mobility and hence the ability of a worker to threaten to leave for a better-paying job.  Noncompete clauses in many labor contracts – including for low-wage workers – may legally block workers from switching to a new employer in the industry where that worker has the particular skills to do well.  The FTC has estimated that 18% of all US workers are covered by noncompete clauses.  The FTC thus approved on April 23, 2024, new regulations banning their use.  While the rule is scheduled to enter into effect on September 4, 2024, it will undoubtedly be challenged in court, with this leading to delays before it can enter into effect (if it ever does).

There is also the separate practice of antipoaching clauses.  These are common in the fast-food industry as well as in other national chains of franchises.  The antipoaching clauses are not in the labor contracts themselves, but rather in the franchise agreements between the franchise owner and the national firm.  They require that the franchise owner not employ any individual who had worked at another franchisee’s establishment sometime before – typically at some point in the prior six months.  McDonald’s claims it ended requiring those clauses in its franchisee contracts in 2017, and several states have banned the practices within their borders.  But McDonald’s is still being sued in court, and it appears the practice remains common.  The new FTC rule – if upheld in court – may apply to these practices as well.

4.  Information is also far from complete nor is it cost-free.  A firm can never know for sure how a particular worker will perform in a job until they are already on the job (with it then costly to fire and replace them in case the performance is not good).  Nor will the worker easily know what all the job opportunities are out there, and what he or she would be paid at some alternative firm.

The relevant information may also be more readily available to one side of the transaction than to the other – what economists call “asymmetric information”.  The worker may know well his or her skills and abilities, but the prospective hiring firm will not.  Similarly, the hiring firm may know well what is needed to do well in a job, but the prospective worker will not.  Also, doing well in a particular job is more than simply a skill set.  It also requires an ability to work well with colleagues and a willingness to take the work seriously.

Firms will thus be cautious in hiring and may only be willing to pay a relatively low wage to new workers to start.  Alternative firms will act similarly, as those firms are also unsure how well a new employee might work out (information is not complete).  Thus they too will only offer a relatively low wage to start.  Plus there are significant costs in the hiring and firing process itself.  All this serves to lock in workers at the firms where they are now, without a credible threat to move elsewhere if their wages are not raised to reflect their full productivity.

5.  Workers also gain firm-specific skills simply by the time they spend at the job.  This spans the range from skills for the specific tasks that the job entails, to understanding better how the firm approaches what they want from those in these jobs, to getting to know colleagues better and their specific likes, dislikes, and how they do things.  These skills are helpful, and lead to the worker becoming more productive at that particular firm.

But while a worker may see his or her productivity rise over time at some particular firm, they will not necessarily see their wage rise by the same amount.  That is, the workers would be paid less than the value of their marginal product.  While the firm might pay the worker somewhat more simply to help lock them in, this would not necessarily reflect the full amount of their higher productivity at that firm.  The worker would not have a credible threat to leave to go to a competing firm where he or she would be paid more.  Their productivity at an alternative firm – where they would once again be starting out – would not be as high and those firms would not be willing to offer a higher wage.

6.  There is also a more fundamental problem in the ability (or rather inability) to ascertain what the productivity is of an individual worker.  One of the assumptions of the neoclassical economic analysis noted above is that the relationship between the input of individual workers and the output of the firm is strictly concave.  That is, as the input of the worker goes up (more hours) there will be a smooth decline in the extra output of the firm as a result of the increased labor input, with no “bumps” in that curve.

Economists call this diminishing marginal returns.  If one increased labor input by a unit, one would see some increase in output.  Increase the labor input by another unit, one would see an increase in output again, but by less than in the first step.  And when the relationship is strictly convex, the increase in output would be less and less for each unit increase in labor input, up to a point where there would be no further increase in output (and after which it might even decline).

Reality is more complex.  Those working in firms are not working simply as individuals but as part of teams.  Adam Smith in the first few pages of The Wealth of Nations in 1776 already noted how far more productive workers can be when working in teams than when trying to do it all individually – the famous pin factory.  It still applies today, and not simply in factories.  Take, for example, a team working a shift at a fast food restaurant.  There may normally be a team of, say, ten for a particular shift.  Each worker has different responsibilities, but most of the workers have the skills to do most or perhaps all of the individual tasks.

In this made-up example, they arrived at a team of ten as normally best to handle a particular shift based on how the tasks can be divided up and given the number of customers they normally expect.  It would be difficult to do with just nine, and not much gained with an extra worker and thus eleven on that shift.

What then is the marginal product of each of the workers?  They need to know this to determine what wages they could pay in the standard neoclassical theory, but it is not well defined.  Starting with any grouping of nine workers, the marginal product from hiring a tenth worker would be relatively high as they then could organize into the optimal team of ten.  But any one of the workers could be considered to be the tenth one added to the team, and hence responsible for the jump in output in going from what is possible with just nine workers to the more productive team of ten.  And if all of the workers were paid a wage corresponding to that jump in output that is possible when going to a full team of ten, they would together be paid more than the overall value of what is being produced with a team of ten.

While the workers would likely welcome such higher wages, the reality is that fast-food restaurants do not aim to operate at a loss.  And they don’t.  Their workers are simply not paid that much.  There are fundamental conceptual problems in trying to define the marginal product of a worker when work takes place in teams (as it normally is).

E.  Final Points and Conclusion 

California has raised its minimum wage repeatedly in the past decade, but there is no indication in the data that this has led to an increase in unemployment.  While economic theory would predict that in “perfectly competitive markets” the workers being paid below the new minimum wage would be laid off (as wages are set, under these assumptions, based on productivity, and they assert that the productivity of such workers is simply too low), this only holds under unrealistic assumptions.  Wage determination is more complex.  In the real-world conditions under which wages are in fact set, it is not a surprise to find that unemployment did not in fact go up.

This does not mean, however, that any increase in the minimum wage would not lead to higher unemployment.  If the minimum wage was set next year at, say, $100 per hour, one should of course expect issues.  What we see in the data is not that there can be any increase in the minimum wage with then no consequences for unemployment, but rather that the increases in the minimum wage that were mandated in California in the last decade did not lead to an increase in the rate of unemployment.

Increases in the minimum wage may also lead to increases in the prices of certain goods.  If the production of those goods were heavily reliant on minimum wage workers, and the firms would now have to pay a higher wage for those workers, it may well be the case that such goods will now only be available at a higher price.  Fast-food hamburgers may go up in price, but don’t view this as simply affecting “junk food”.  The prices of blueberries and strawberries might go up as well.

Does this mean that the critics of the minimum wage are in fact right?  No, it does not.  First, it remains the case that unemployment did not go up following the major increases in the minimum wage in California over the past decade.  The critics asserted that it would.

Second, while prices of fast-food hamburgers may have gone up following the increases in the minimum wage, those prices did not go up by as much as the minimum wage did.  If wages in fact reflected the value of the marginal product of the worker, the wages of the minimum wage workers would still have gone up relative to that value – just not by as much.  Under this theory of wage determination, they would still have been laid off.  But there is no evidence of this in the data.

Labor markets operate far from what economists would call “perfectly”.  In this reality, minimum wage laws can play a valuable and indeed important role.